Browse All Artificial Intelligence Content (845)

Allison announces the June 1, 2026 deprecation of GPT-5.2 and GPT-5.2-Codex across GitHub Copilot experiences, and outlines the replacement models plus the admin steps needed to ensure the alternatives are available via Copilot model policies.

mosiddi explains how the Agent Governance Toolkit (AGT) “shifts left” governance for AI agents by catching security and compliance violations before runtime, using pre-commit hooks, PR gates, CI checks, and release-time controls like SBOMs, signing, and provenance attestations.

John Edward discusses how GitHub Copilot changes programming education, where it can speed up learning, and where it can undermine fundamentals if students rely on it too heavily. The post outlines practical habits for students and classroom approaches for educators to use Copilot without losing academic rigor.

John Maeda shares a practical tip from his SXSW “Design In Tech” report: every generative AI workflow needs a “standing beacon” to keep outputs consistent. He explains why evaluation, guardrails, and a clear standard (a “judge” that can reject bad generations) are essential when iterating on AI-generated work.

Nirav Shah, Rob Lefferts and Jason Roszak announce the general availability of Microsoft Agent 365 and outline new security and governance capabilities for discovering, inventorying, and controlling AI agents across endpoints, SaaS, and cloud platforms, including Defender/Intune-based shadow agent discovery and Entra network controls.

Kirill Gavrylyuk hosts a Cosmos DB Conf 2026 keynote panel with leaders from Vercel, OpenAI, Microsoft, Walmart, and AMD on how modern, high-demand applications are built and scaled on Azure Cosmos DB, and how AI and data platform needs are shaping the next wave of distributed systems.

John Savill runs through the 1st May 2026 Azure update, covering new and preview features across AKS networking and troubleshooting, Azure Front Door WAF protections, Elastic SAN improvements, PostgreSQL read replica scaling, and major AI platform changes like Microsoft Agent Framework GA and Prompt flow retirement.

HimanshuYadav explains how to modernize brownfield Terraform codebases by refactoring legacy modules to Azure Verified Modules (AVM) with AI assistance. The post focuses on using tools like GitHub Copilot to draft changes, then relying on disciplined Terraform plan review and policy gates to keep state changes safe.

Ravindra Kumar Vishwakarma explains how GitHub Copilot CLI can run as an Agent Client Protocol (ACP) server, enabling tools, IDEs, and CI/CD systems to connect to Copilot as a backend agent with streaming, sessions, and permissioned tool execution.

Kristen Womack summarizes the April 2026 Azure Developer CLI (azd) releases, focusing on multi-language hooks in azure.yaml, extension framework updates, Copilot-assisted troubleshooting, and a set of security, provisioning, and CI/CD reliability improvements across versions 1.23.14 through 1.24.2.

Rob Bos breaks down five GitHub Copilot and agent extensibility surfaces that create supply-chain and governance gaps in large enterprises, and explains what controls exist today (and where they don’t) across Copilot CLI plugins, APM, gh skill, MCP servers, and VS Code extension registries.

The Visual Studio Code Team summarizes what’s new in VS Code 1.119 (Insiders), including Markdown UX updates and several chat/agent features such as improved codebase context for virtual file systems, attaching browser tabs as context, and Copilot CLI plan mode support.

stclarke summarizes recent Microsoft Security updates, including new Microsoft Defender protections for AI agents in the Agent 365 tooling gateway, a generally available Defender for Cloud + GitHub Advanced Security integration, and a new Microsoft Purview demo for running data security investigations with AI-assisted analysis.

Luis Quintanilla walks through ConferencePulse, a Blazor Server app on .NET 10 that uses .NET’s composable AI stack—Microsoft.Extensions.AI, DataIngestion, VectorData, MCP, and Microsoft Agent Framework—to build RAG-based Q&A, AI-generated polls/insights, and a multi-agent session summary.

Kayla Cinnamon explains how GitHub Copilot CLI’s interactive and non-interactive modes differ, when to use each, and the exact commands to start a session, run one-off prompts, and resume prior sessions from the terminal.

GitHub demonstrates how hooks work in the GitHub Copilot CLI, explaining the agent lifecycle and when hook events fire (session start, prompt submitted, and pre/post tool use). The video shows how to use these events to influence tool calls and tailor Copilot CLI behavior to your workflow.

Allison summarizes the April 2026 update for GitHub Copilot in Visual Studio 2026, focusing on agentic workflows like launching cloud agent sessions from the IDE, user-level custom agents and skills discovery, a new debugger agent flow, plus chat history, shortcut customization, and new C++ tools.

Microsoft Threat Intelligence and Microsoft Defender Security Research Team summarizes Q1 2026 email threat trends, including the scale of phishing, the surge in QR code and CAPTCHA-gated campaigns, and how the Tycoon2FA disruption shifted attacker infrastructure. It also lists concrete Defender/EOP mitigations and relevant detections.

Microsoft Fabric Blog announces a preview of the OneLake Catalog Search REST API plus MCP and Fabric CLI support, letting developers search Fabric items across workspaces by metadata and use the returned identifiers in scripts, internal tools, and agent-driven workflows.

Microsoft Fabric Blog explains how Fabric SQL developers can move from Azure Data Studio to VS Code, keeping SQL Database Projects and adopting a Git-based workflow with pull requests, schema compare, publish script previews, and optional GitHub Copilot assistance in the MSSQL extension.